I'll bet you're wondering what it is that might come here…. Who uses crappy computers to solve big problems?

Well, the answer is, I do.

I just graduated with a BS in CS from GU, and I've been doing research for the last year in AI and Parallel Processing. For the AI, I've been studying GA and Neural Nets and am working up a new way to look at combining the two (my "Big Problem"). I realized this was to be hugely computationially intensive from the get-go, so I got interested in parallel processing, and involved myself in a clustering project on campus. We had 20-ish 800mhz machines and roughly the same number of 333mhz machines (my "Crappy Computers") to work with, and did some really interesting work.

The point of this blog is to review some of that research, discuss some of my thought processes in doing what I did, and to continue documenting my research. I've never been one for academic writing, yet I feel like my research deserves some credit, so why not blog, eh? This process is going to start out a little slow, as I a: don't have internet at home yet, and b: am in the process of building my own cluster since the university didn't see fit to let a lowly undergrad snag a few of the machines he meticulously worked on all semester post-graduation, but I digress… The point is, be patient, and I will show you great things.

I should let you know now, I have a habit of coming up with metaphors, and more than likely this blog will be full of them. Perhaps they'll be entertaining, but let's not get ahead of ourselves.

Advertisements

5 Responses to “Why read this?”

  1. Ryan Says:

    I thought I’d leave a commment so you’d have at least one. The kids here are working parallelizing an NP-Complete problem too but they are making an internet game that when people play it ‘steals’ some of their processing time to use in their algorithm. There’s no way they can finish it but its kind of cool. Anyway have fun with the parallelizing!


  2. Hi, I’m interested in your work. I graduated long ago – way before Beowolf/MPI/PVM or openMosix PC based clusters. However, as an AP Computer Science teacher, I’d like to learn a bit about clusters and hopefully share something with my students. I’m wondering what you would recommend? My students know java, but java clusters don’t seem ready for prime time (except perhaps parallel java). So, I’ve been looking at the BCCD project which makes a live linux CD with MPI, openMosix and some apps on it. I simply boot a number of CDs and thats how many nodes I have in a PC based openMosix cluster. What I would like is an easy way to program them and do some number crunching. I see that you are testing primes, how did you do that? Is this like the Great Internet Mersene Prime Search? I’ve written some c++ code that seems to migrate using fork()/wait() and I’ve run povray, but that’s about it. I’d like to do a mersenne search using c++ objects to simulate very large integers. I’d also like to do some mandelbrot graphics. What do you think? TIA, AJG

  3. Boobs Says:

    Is athlon dual core 2.8 ghz considered a crappy computer?


  4. Athlon is a good processor Boobs. You should get it.


  5. woops, i meant it was crappy. Click on my name and ull see why

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s